Main
Petr Šimeček
Data Scientist, Bioinformatics Analyst, ML Engineer
Professional Experience
Biostatistician
Institute of Animal Science
Prague, Czechia
2007 - 2009
- Designing experiments
- Categorical data analysis
- Mixed-effects models
- GPS tracking data
Bioinformatician
Institute of Molecular Genetics
Prague, Czechia
2007 - 2017
- Mouse genetics
- Next generation sequencing
- Metabolomics
- Later Head of Bioinformatics Unit
- IMPC database
Bioinformatics Analyst
The Jackson Laboratory
Bar Harbor, Maine, USA
2013 - 2017
- QTL mapping
- Mouse diversity outbred
- Mediation analysis
- Aging and its effect on proteome
- R/Shiny & Docker
Data Scientist
Google LLC
Mountain View, California, USA
2017 - 2018
- Time Series: development and maintenance of internal time series forecasting tool
- Various ad hoc analysis
- Deep learning applied to time series forecasting
Machine Learning Engineer
Central European AI Institute (CEAi)
Brno, Czechia
2019
- ML model to predict house prices
- Gradient boosting (XGBoost, LightGBM, CatBoost), neural networks (fast.ai, keras, TF)
- Amazon EC2, S3, ECS, Elastic Beanstalk, CloudWatch, Apache Airflow
Teaching And Selected Talks
Introduction to R Language for Beginners.
Instructor of Software Carpentry and Software for Scientists, https://crabhi.github.io/2016-10-08-umg/.
Boston, USA & Prague, Czechia
2015 - 2017
Deep Learning: From Zero To Hero in Two Hours.
Workshop with intro to deep learning (together with Karla Fejfarova), https://github.com/simecek/from0toheroin2h.
Prague, Czechia
2018 - 2019
Statistical vs. Deep Learning Methods for Time Series Forecasting.
Recent talk at Machine Learning Meetup, https://youtu.be/mqYwy5RuSQQ
Brno, Czechia
2019
Education
Charles University in Prague
Mgr. (M.Sc.) in Probability Theory and Stochastic Processes (1st prize in the diploma–thesis competition at Department of Probability and Mathematical Statistics in July 2003)
Prague, Czechia
1998 - 2003
Thesis: On the Minimal Probability of Intersection of Dependent Events
Vrije Universiteit
Socrates / Erasmus Exchange
Amsterdam, Netherlands
2002
Hasselt Universiteit
Master of Science in Biostatistics
Hasselt, Belgium
- AIA Fellowship (one of two annually awarded to Czech students)
- MSc. degree with the great distinction
2004 - 2005
Thesis: Gene Expression Data Analysis for In Vitro Toxicology
Charles University in Prague
Ph.D. in Mathematical Statistics and Probability Theory (thesis summary at http://bit.ly/2SazFPc)
Prague, Czechia
2003 - 2007
Thesis: Independence Models
Selected Publications
See my Google Scholar profile for the full list of 20+ papers and >750 citations.
Genetic analysis of substrain divergence in non-obese diabetic (NOD) mice.
G3: Genes, Genomes, Genetics. 2015 May 1;5(5):771-5.
N/A
2015
Simecek P, Churchill GA, Yang H, Rowe LB, Herberg L, Serreze DV, Leiter EH.
Defining the consequences of genetic variation on a proteome-wide scale.
Nature. 2016 Jun;534(7608):500.
N/A
2016
Chick JM, Munger SC, Simecek P, Huttlin EL, Choi K, Gatti DM, Raghupathy N, Svenson KL, Churchill GA, Gygi SP.
High-resolution maps of mouse reference populations.
G3: Genes, Genomes, Genetics. 2017 Oct 1;7(10):3427-34.
N/A
2017
Simecek P, Forejt J, Williams RW, Shiroishi T, Takada T, Lu L, Johnson TE, Bennett B, Deschepper CF, Scott-Boyer MP, de Villena FP.